/root/doris/be/src/util/load_util.cpp
Line | Count | Source |
1 | | // Licensed to the Apache Software Foundation (ASF) under one |
2 | | // or more contributor license agreements. See the NOTICE file |
3 | | // distributed with this work for additional information |
4 | | // regarding copyright ownership. The ASF licenses this file |
5 | | // to you under the Apache License, Version 2.0 (the |
6 | | // "License"); you may not use this file except in compliance |
7 | | // with the License. You may obtain a copy of the License at |
8 | | // |
9 | | // http://www.apache.org/licenses/LICENSE-2.0 |
10 | | // |
11 | | // Unless required by applicable law or agreed to in writing, |
12 | | // software distributed under the License is distributed on an |
13 | | // "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY |
14 | | // KIND, either express or implied. See the License for the |
15 | | // specific language governing permissions and limitations |
16 | | // under the License. |
17 | | |
18 | | #include "util/load_util.h" |
19 | | |
20 | | #include <string> |
21 | | |
22 | | #include "util/string_util.h" |
23 | | |
24 | | namespace doris { |
25 | | void LoadUtil::parse_format(const std::string& format_str, const std::string& compress_type_str, |
26 | | TFileFormatType::type* format_type, |
27 | 60 | TFileCompressType::type* compress_type) { |
28 | 60 | if (format_str.empty()) { |
29 | 3 | parse_format("CSV", compress_type_str, format_type, compress_type); |
30 | 3 | return; |
31 | 3 | } |
32 | | |
33 | | // Assign compress_type |
34 | 57 | if (iequal(compress_type_str, "GZ")) { |
35 | 6 | *compress_type = TFileCompressType::GZ; |
36 | 51 | } else if (iequal(compress_type_str, "LZO")) { |
37 | 3 | *compress_type = TFileCompressType::LZO; |
38 | 48 | } else if (iequal(compress_type_str, "BZ2")) { |
39 | 6 | *compress_type = TFileCompressType::BZ2; |
40 | 42 | } else if (iequal(compress_type_str, "LZ4")) { |
41 | 6 | *compress_type = TFileCompressType::LZ4FRAME; |
42 | 36 | } else if (iequal(compress_type_str, "LZ4_BLOCK")) { |
43 | 3 | *compress_type = TFileCompressType::LZ4BLOCK; |
44 | 33 | } else if (iequal(compress_type_str, "LZOP")) { |
45 | 9 | *compress_type = TFileCompressType::LZO; |
46 | 24 | } else if (iequal(compress_type_str, "SNAPPY_BLOCK")) { |
47 | 3 | *compress_type = TFileCompressType::SNAPPYBLOCK; |
48 | 21 | } else if (iequal(compress_type_str, "DEFLATE")) { |
49 | 6 | *compress_type = TFileCompressType::DEFLATE; |
50 | 15 | } else { |
51 | 15 | *compress_type = TFileCompressType::PLAIN; |
52 | 15 | } |
53 | | |
54 | | // Assign format_type |
55 | 57 | *format_type = TFileFormatType::FORMAT_UNKNOWN; |
56 | 57 | if (iequal(format_str, "CSV")) { |
57 | 21 | if (compress_type_str.empty()) { |
58 | 3 | *format_type = TFileFormatType::FORMAT_CSV_PLAIN; |
59 | 18 | } else if (iequal(compress_type_str, "GZ")) { |
60 | 3 | *format_type = TFileFormatType::FORMAT_CSV_GZ; |
61 | 15 | } else if (iequal(compress_type_str, "LZO")) { |
62 | 3 | *format_type = TFileFormatType::FORMAT_CSV_LZO; |
63 | 12 | } else if (iequal(compress_type_str, "BZ2")) { |
64 | 3 | *format_type = TFileFormatType::FORMAT_CSV_BZ2; |
65 | 9 | } else if (iequal(compress_type_str, "LZ4")) { |
66 | 3 | *format_type = TFileFormatType::FORMAT_CSV_LZ4FRAME; |
67 | 6 | } else if (iequal(compress_type_str, "LZ4_BLOCK")) { |
68 | 0 | *format_type = TFileFormatType::FORMAT_CSV_LZ4BLOCK; |
69 | 6 | } else if (iequal(compress_type_str, "LZOP")) { |
70 | 3 | *format_type = TFileFormatType::FORMAT_CSV_LZOP; |
71 | 3 | } else if (iequal(compress_type_str, "SNAPPY_BLOCK")) { |
72 | 0 | *format_type = TFileFormatType::FORMAT_CSV_SNAPPYBLOCK; |
73 | 3 | } else if (iequal(compress_type_str, "DEFLATE")) { |
74 | 3 | *format_type = TFileFormatType::FORMAT_CSV_DEFLATE; |
75 | 3 | } |
76 | 36 | } else if (iequal(format_str, "HIVE_TEXT")) { |
77 | 0 | *format_type = TFileFormatType::FORMAT_TEXT; |
78 | 36 | } else if (iequal(format_str, "JSON")) { |
79 | 30 | *format_type = TFileFormatType::FORMAT_JSON; |
80 | 30 | } else if (iequal(format_str, "PARQUET")) { |
81 | 3 | *format_type = TFileFormatType::FORMAT_PARQUET; |
82 | 3 | } else if (iequal(format_str, "ORC")) { |
83 | 3 | *format_type = TFileFormatType::FORMAT_ORC; |
84 | 3 | } else if (iequal(format_str, "WAL")) { |
85 | 0 | *format_type = TFileFormatType::FORMAT_WAL; |
86 | 0 | } else if (iequal(format_str, "ARROW")) { |
87 | 0 | *format_type = TFileFormatType::FORMAT_ARROW; |
88 | 0 | } |
89 | 57 | } |
90 | | |
91 | 42 | bool LoadUtil::is_format_support_streaming(TFileFormatType::type format) { |
92 | 42 | switch (format) { |
93 | 3 | case TFileFormatType::FORMAT_CSV_PLAIN: |
94 | 6 | case TFileFormatType::FORMAT_CSV_BZ2: |
95 | 9 | case TFileFormatType::FORMAT_CSV_DEFLATE: |
96 | 12 | case TFileFormatType::FORMAT_CSV_GZ: |
97 | 15 | case TFileFormatType::FORMAT_CSV_LZ4FRAME: |
98 | 15 | case TFileFormatType::FORMAT_CSV_LZ4BLOCK: |
99 | 18 | case TFileFormatType::FORMAT_CSV_LZO: |
100 | 21 | case TFileFormatType::FORMAT_CSV_LZOP: |
101 | 24 | case TFileFormatType::FORMAT_JSON: |
102 | 24 | case TFileFormatType::FORMAT_TEXT: |
103 | 24 | case TFileFormatType::FORMAT_WAL: |
104 | 24 | case TFileFormatType::FORMAT_ARROW: |
105 | 24 | return true; |
106 | 18 | default: |
107 | 18 | return false; |
108 | 42 | } |
109 | 0 | return false; |
110 | 42 | } |
111 | | } // namespace doris |